This article has been accepted for publication in a future issue of this journal, but has not been

fully edited. Content may change prior to final publication.

Future on Power Electronics for Wind Turbine Systems

Frede Blaabjerg and Ke Ma
Department of Energy Technology, Aalborg University, Aalborg DK-9220, Denmark fbl@et.aau.dk, kema@et.aau.dk
ABSTRACT - Wind power is still the most promising renewable energy in the year of 2013. The Wind Turbine System started with a few tens of kW power in the 1980s, nowadays multi-MW wind turbines are widely installed even up to 6-8 MW. There is a wide-spread use of wind turbines in the distribution networks and more and more wind power stations, acting as power plants, are connected directly to the transmission networks. As the grid penetration and power level of the wind turbines increase steadily, the wind power start to have significant impacts to the power grid system. Therefore more advanced generator, power electronic system, as well as control solutions have to be introduced to improve the characteristics of the wind power plant and make it more suitable to be integrated into the power grid. Meanwhile there are also some emerging technology challenges which need to be further clarified and investigated. This paper gives an overview and discusses some development trends in the technologies used for wind power systems. First, the developments of technology and market are generally discussed. Next several state-of-the-art wind turbine concepts, as well as the corresponding power electronic converters and control structures are reviewed respectively. Furthermore the grid requirements as well as the technology challenges for the future wind turbine system are also addressed.

but also enable the whole WTS to act like a completely controllable generation unit being able to much better integrate the wind power into the power grid [2]-[6], [34]-[37]. The scope of this paper is to give a status overview and discuss some technology trends of power electronics used for wind power application. First, the technology and market developments of wind power generation are generally introduced. Next two dominant wind turbine concepts as well as the potential converter topologies are addressed. Then some control methods and demands are briefly explained for the state-of-the art wind turbines. Further, the emerging challenges for future wind power generation are discussed and some final conclusions are given. II. DEVELOPMENT OF WIND POWER GENERATION The cumulative wind power capacity from 1999 to 2020 is shown in Fig. 1, and it can be seen that the wind power has grown fast to a capacity of 283 GW with around 45 GW installed only in 2012, and this number is expected to achieve 760 GW in 2020 on moderate scenario [9]. The wind power grows more significant than any other renewable energy sources and is becoming really an important player in the modern energy supply system. As an extreme example, Denmark has a high penetration by wind power and today more than 30 % of the electric power consumption is covered by wind. This country has even the ambition to achieve 100 % non-fossil-based power generation system by 2050 [10].

I. INTRODUCTION As the fast growing capacity and more significant impacts to the power grid by Wind Turbine System (WTS), the power electronic technologies used in wind power application have changed dramatically during the last 30 years [1]-[18]. In the 1980s, the power electronics for wind turbines was just a softstarter used to initially interconnect the Squirrel-Cage Induction Generator (SCIG) with the power grid, and only simple thysistors were applied and they did not need to carry the power continuously [2]. In the 1990s the power electronic technology was mainly used for the rotor resistance control of Wound-Rotor Induction Generator (WRIG), where more advanced diode bridges with a chopper were used to control the rotor resistance for generator [4], especially at nominal power operation in order to reduce mechanical stress and loading. Since 2000, even more advanced back-to-back power converters were introduced in large scale, which started to regulate the generated power from the wind turbines: First mostly in the partial-scale power capacity for the Doubly Fed Induction Generator (DFIG) [11]-[15], then in the full-scale power capacity for the Asynchronous/Synchronous Generator (A/SG) [1], [6], [28]-[33]. By introducing the back-to-back power electronics converters, it is possible to fully control the extracted power from the wind turbines, and also provide ancillary services to the grid. Power electronics gradually become more and more advanced and bring in significant performance improvements for the wind turbines - not only reducing the mechanical stress and increase the energy yield,

Fig. 1. Global cumulative installed wind power capacity from 1999 to 2020.

Regarding to the markets and manufacturers of wind power, China is still the largest market with over 13 GW capacity installed in 2012, together with the USA (13.1 GW) and the EU (11.9 GW) sharing around 87 % of the global market. The Danish company Vestas first gives out the top position among the largest manufacturers since 2000, while GE catches up to the first due to the strong U.S. market in 2012. Fig. 2 summarizes the worldwide top suppliers of wind

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

turbines in 2012. It is can see that there are four Chinese companies in the Top 10 manufacturers with a total market share of 16.6 %, which is a significant drop compared to the 26 % in 2011 [9].

designs can generally be categorized into several concepts [3][5]. In these wind turbine concepts the power electronics play quite different role in the WTS and has various power rating coverage of the system, as shown in Fig. 3. Up until now the configuration of Doubly Fed Induction Generator (DFIG) equipped with partial-scale power converter is dominating on the market, but in very near future the configuration with Synchronous Generator (SG) with full-scale power converter is expected to take over. Actually the solutions with full-scale power converter are becoming the preferred technology choices in the bestselling power ranges of the wind turbines [6], [7]. In the following these two state-of-the-art wind turbine concepts are going to be introduced. A. Doubly-Fed Induction-Generator with partial-scale power converter This wind turbine concept is the most adopted solution nowadays and it has been used extensively since 2000s. As shown in Fig. 4, a power electronics converter is adopted in conjunction with the Doubly-Fed Induction Generator (DFIG). The stator windings of DFIG are directly connected to the power grid, while the rotor windings are connected to the power grid by the converter with normally 30% capacity of the wind turbine [11], [12]. In this concept, the frequency and the current in the rotor can be flexibly regulated and thus the variable speed range can be extended to a satisfactory level. The smaller converter capacity makes this concept attractive seen from a cost point of view. However, its main drawbacks are the use of slip-rings and the challenging power controllability in the case of grid faults these disadvantages may comprise the reliability and may be difficult to completely satisfy the future grid requirements as claimed in [13], [14].

Fig. 2. Distribution of wind turbine market share by manufacturers in 2012 [9].

Besides the quick growth in the total installed capacity, the size of individual wind turbine is also increasing dramatically in order to obtain a reduced price per generated kWh. In 2012 the average turbine size delivered to the market was 1.8 MW, among which the average offshore turbine has achieved a size of 4 MW. The growing trends of emerging turbine size between 1980 and 2018 are shown in Fig. 3, where the development of power electronics in the WTS (rating coverage and function role) is also indicated. It is noted that the cutting-edge 8 MW wind turbines with diameter of 164 m have already shown up in 2012 [7]. Right now most turbine manufacturers are developing products in the power range of 4.5-8 MW, and it is expected that more and more large wind turbines with multi-MW power level, (even up to 10 MW will appear in 2018), will be present in the next decade - driven mainly by the considerations to lower down the cost of energy [9].

Fig. 4. Variable speed wind turbine with partial-scale power converter and a doubly-fed induction generator.

Fig. 3. Evolution of wind turbine size and the power electronics seen from 1980 to 2018 (Estimated), blue circle indicates the power coverage by power electronics.

III.

STATE-OF-THE-ART CONFIGURATIONS AND POWER

ELECTRONICS FOR WIND TURBINES

The Two-Level Pulse-Width-Modulation Voltage-SourceConverter (2L-PWM-VSC) is the most used converter topology so far for the DIFG based wind turbine concept as the power rating requirement for the converter is limited. Normally, two 2L-PWM-VSCs are configured in a back-toback structure in the wind turbine system, as shown in Fig. 5 which is called 2L-BTB for convenience. A technical advantage of the 2L-BTB solution is the full power controllability (4-quadrant operation) with a relatively simple structure and few components, which contribute to wellproven robust/reliable performances as well as advantage of cost [15].

The development of the wind turbine technology has been steady for the last 35 years [1]-[6]. Depending on the types of generator, power electronics, speed controllability and the way in which the aerodynamic power is limited, the wind turbine

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

industry for the wind turbines with power level higher than 3 MW.

Fig. 5. Two-level back-to-back voltage source converter for wind turbine (2L-BTB).

B. Asynchronous/synchronous generator with full-scale power converter The second important concept that is popular for the newly developed and installed wind turbines is shown in Fig. 6. It introduces a full-scale power converter to interconnect the power grid and stator windings of generator, thus all the generated power from the wind turbine can be regulated. The Asynchronous Generator, Wound Rotor Synchronous Generator (WRSG) or Permanent Magnet Synchronous Generator (PMSG) have been reported as solutions to be used. The elimination of slip rings, simpler or even eliminated gearbox, full power and speed controllability as well as better grid support ability are the main advantages compared to the DFIG based concept. However, the more stressed and expensive power electronic components as well as the higher power losses in the converter are the main drawbacks for this concept.

(a) With multi-winding generator

(b) With regular winding generator Fig. 7. Multi-cell converter with paralleled 2L-BTB converter cells.

Fig. 6. Variable speed wind turbine with full-scale power converter.

As the converter in this concept needs to carry all the generated power by wind turbines e.g. up to 10 MW, the 2LBTB converter topology at this power level may suffer from large switching loss and many devices may need to be connected in parallel, also the cabling in case of low voltage level can be a great design/physical challenge. Consequently, it is becoming more and more difficult for a single 2L-BTB topology to achieve acceptable performance for the full-scale wind power converter, even though having the cost advantage. In order to handle the growing power with the exiting 2LBTB technology, some multi-cell converter configurations are introduced (i.e. parallel/series connection of 2L-BTB converter cells). Fig. 7 (a) shows a multi-cell solution adopted by Gamesa in the 4.5 MW wind turbines [16], which have several 2L-BTB converters paralleled both on the generator side and on the grid side. Siemens also introduce the similar solution in their best-selling multi-MW wind turbines, as indicated in Fig. 7 (b) [17]. The standard and proven low voltage converter technologies as well as redundant and modular characteristics are the main advantages. This converter configuration is the state-of-the-art solution in the

With the abilities of achieving higher voltage and power level, multi-level converters may become more preferred candidates in the full-scale converter based concept [18]-[20]. The Three-level Neutral Point diode Clamped topology is one of the most commercialized multi-level topologies on the market. Similar to the 2L-BTB, it is usually configured as a back-to-back structure in the wind power application, as shown in Fig. 8, which is called 3L-NPC BTB for convenience. The 3L-BTB solution achieves one more output voltage level and less dv/dt stress compared to the counterpart of 2L-BTB, thus it is possible to convert the power at medium voltage with lower current, less paralleled devices, and smaller filter size. The mid-point voltage fluctuation of the DC-bus can be a drawback of the 3L-NPC BTB. This problem has been extensively researched and it is considered to be improved by the controlling of redundant switching states [21]-[23]. However, it is found that the loss distribution is unequal between the outer and inner switching devices in a switching arm [24]-[26], and this problem might lead to a derated power capacity, when it is practically used.

Fig. 8. Three-level Neutral Point Clamped converter for wind turbine (3L-NPC BTB).

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

IV.

CONTROL STRUCTURE OF WIND TURBINE SYSTEM

Controlling a wind turbine involves both fast and slow control dynamics [27]-[33], as indicated in Fig. 9, where a general control structure for a wind turbine system including turbine, generator, filter and converter are indicated. The wind turbine concept can either be the type shown in Fig. 4, or the type shown in Fig. 6. Generally the power flowing in and out of the generation system has to be managed carefully. The generated power by the turbines should be controlled by means of mechanical parts (e.g. pitch angle of blades, yawing system, etc.). Meanwhile, the whole control system has to

follow the power production commands given by Distribution/Transmission System Operator (DSO/TSO). More advanced features of the wind turbine control may be taken into account like the maximization of the generated power, ride through operation of the grid faults and providing grid supporting functions in both normal and abnormal operations, etc. In the variable speed wind turbine concept, the current in the generator will typically be changed by controlling the generator side converter, and thereby the rotational speed of turbine can be adjusted to achieve maximum power production based on the available wind power. In respect to operation under grid fault, coordinated

Turbine Gear
DFIG AC S DC SG/PMSG I IG

Chopper

AC DC

Filter

Grid

Vdc IGen.
gen.

Pulse Width Modulation (PWM) Grid Current/Voltage Synchronization Control Basic control loops Power Maximization & Limitation Fault ride through & Grid support

Xfilter Igrid Vgrid Pmeas.,Qmeas.

Command from TSO

WT specific controls Inertia Emulation Energy Storage Power Quality

Extra control features

Fig. 9. General control structure for modern wind turbines.

Fig. 10. Control of a wind turbine with doubly-fed induction generator.

Fig. 11. Control of active and reactive power in a wind turbine with multi-pole synchronous generator.

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

control of several subsystems in the wind turbine like the generator/grid side converters, braking chopper/crowbar, pitch angle controller is necessary. Finally, the basic controls like current regulation, DC bus stabilization and the grid synchronization have to be quickly performed by the wind power converter, where the Proportional-Integral (PI) controller and ProportionalResonant (PR) controllers are typically used [31], [34], [35]. As an example, the control methods for a DFIG based wind turbine system are shown in Fig. 10. Below maximum power production the wind turbine will typically vary the rotational speed proportional with the wind speed and keep the pitch angle fixed. At very low wind speed the rotational speed will be fixed at the maximum allowable slip in order prevent over voltage of generator output. A pitch angle controller is used to limit the power when the turbine output above the nominal power. The total electrical power of the wind turbine system is regulated by controlling the DFIG through the rotor-side converter. The control strategy of the grid-side converter is simply just to keep the dc-link voltage fixed. It is noted that a trend is to use a crowbar connected to the rotor of DFIG in order to improve the control performance under grid faults [13], [14]. Another example for the control structure used for fullscale converter based wind turbine concept is shown in Fig. 11. An advantage of this turbine system is that the dc-link performs some kind of control decoupling between the turbine and the grid. The dc-link will also give an option for the wind turbines to be connected with energy storage units which can better manage the active power flow into the grid system this feature will further improve the grid supporting abilities of the wind turbines. The generated active power of the wind turbine system is controlled by the generator side converter, while the reactive power is controlled by the grid side converter. It is noted that a DC chopper is normally introduced to prevent over voltage of DC link in case of grid faults, when the extra turbine power needs to be dissipated as the sudden drop of grid voltage. V. GRID INTEGRATION OF WIND TURBINE SYSTEM The fluctuation and unpredictable features of wind energy are non-preferred for grid operation. Most countries have strict requirements for the behaviour of wind turbines, known as grid codes which are updated regularly [36]-[40]. Basically, the grid codes are always trying to make the WTS to act as a conventional power plant from the electrical utility point of view. That means the WTS should not only be a passive power source simply injecting available power from the wind, but also behave like an active generation unit, which can wisely manage the delivered active/reactive power according to the demands, and provide frequency/voltage support for the power grid. Examples of the state-of-the-art grid supporting requirements are given in the following. They are specified either for individual wind turbine or for the whole wind farm. According to most grid codes, the individual wind turbines must be able to control the active power at the Point-ofCommon-Coupling (PCC). Normally, the active power has to be regulated based on the grid frequency, e.g. in Denmark,

Available power
100% 75%

With full production

50% 25%

With reduced production

fg (Hz)
48
48.7

49

50
49.85 50.15

51
51.3

52

Fig. 12. Frequency control profiles for the wind turbines connected to the Danish grid [39].

Fig. 13. Reactive power ranges under different generating powers for a wind farm specified by the German grid codes [40].

Fig. 14. Voltage profile for low voltage ride-through capability of wind turbines by different contries [37].
Iq /Irated
100%

Dead band

0 0.5

Vg (p.u.)
0.9 1.0

Fig. 15. Reactive current requirements for a wind farm during grid sags by the German and Danish grid code [39], [40].

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

Table I. Comparison of grid integration performances between conventional power plant and wind turbine plant, the more + the better, 0 means moderate, - means no such ability, PEC means Power Electronics Converter.

Grid integration Features Active power control Reactive power control Short circuit capability Voltage backup Power output Inertia Frequency control Black start capability

Conventional power plant

WTG in the past (without/few PEC)

WTG nowadays/future (With PEC)

+ + ++ ++ ++ ++ +

0 0/+ -

+ ++ ++ + + ++ +

Ireland and Germany, so that the grid frequency can be somehow maintained. As an example the characteristic of frequency supports in the Danish grid codes is shown in Fig. 12, where the active power should be decreased when the frequency rises above 48.7 Hz or 50.15 Hz depending on the power reserving strategy [39]. Similarly, the reactive power delivered by the WTS also has to be regulated in a certain range. As shown in Fig. 13 both the Danish and German grid codes give a range of the reactive power delivered by the WTS to the active power output [40] - this will lead to larger MVA capacity when designing the whole converter system. Also the Transmission System Operator (TSO) will normally specify the reactive power range of the wind turbine system according to the grid voltage levels. It is noted that this reactive power control should be realized slowly under the time constant of minutes in steady state operation [36]. Besides the normal operation, the TSOs in different countries have issued strict grid supporting requirements for the WTS under grid faults. As shown in Fig. 14 [36], [37], in which the boundaries with various grid voltage dip amplitudes as well as the allowable disturbing time are defined for a wind farm. It is becoming a need that the WTS should also provide reactive power (up to 100% current capacity) to contribute to the voltage recovery, when the grid voltage sag is present. Fig. 15 shows the required amount of reactive current against to the grid voltage amplitude by the German [40] and Danish grid codes [39]. This demand is relatively difficult to be met by the wind turbine concept in Fig. 4, and other power quality units like STATCOMs may probably be introduced to help the wind turbine system achieve this tough requirement. The requirements for more grid supports by wind turbines on one hand have increased the cost per produced kWh, but on the other hand made the wind energy more suitable to be largely utilized and integrated into the power grid. It can be predicted that the stricter grid codes in the future will keep challenging the wind turbine system and pushing forward the power electronic technologies. Table I compares the features of conventional power plant, Wind Turbine Generation (WTG) system in the past and WTG system nowadays, where the

major grid integration performances are focused. It can be concluded that by introducing more advanced power electronics, controls and grid regulations, the start-of-the-art wind turbine system equipped with synchronous generator and full-scale power converter can more or less emulate the behaviours of conventional power plants, making the wind power technology to be more suitable integrated into the power grid. VI. TECHNOLOGY CHALLENGES FOR POWER ELECTRONICS IN
WIND TURBINE SYSTEM

In this section several emerging technology challenges for the power electronics in the future wind turbine system are addressed. The discussions will mainly focus on technology issues of power electronics in respect to cost, reliability, grid integration, new power electronics circuits and devices. A. Lower Levelized Cost of Energy (LCOE) The cost issues are the most important considerations for the technology which will determine the feasibility of certain energy technologies to be widely utilized in the future. The competitive cost advantage is the main reason why the wind power showed the most significant growth in the last few decades compared to the other renewable energy sources. In order to quantify and compare the cost for different energy technologies, Levelized Cost of Energy (LCOE) index is generally used [41]. LCOE represents the price at which the electricity is generated from a specific energy source over the whole lifetime of the generation unit. It is an economic assessment of the cost of the energy-generating system including all the costs like: initial investment, development cost, capital cost, operations and maintenance cost, cost of fuel, etc. LCOE can be defined in a single formula as:

LCOE

CDev CCap CO&M E Annual

(1)

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

It is noted that the initial development cost CDev, capital cost Ccap, and the cost for operation and maintenance CO&M are first levelized to annual average cost seen over the lifetime of the generation system, and then divided by the average annual energy production in the whole lifetime EAnual. Seeing from (1) it is clear that in order to reduce the cost of energy, one effective way is to reduce the cost for development, capital, operation and maintenance, the other effective way is to extend the energy production or increase the lifetime of the generation system. As an example the LCOE for offshore wind power of Denmark and U. K. is between 140 EUR/MWh and 180 EUR/MWh in 2010 according to the studies carried by [42], and this number is expected to reduce 50% by 2020 in the span of 67 EUR/MWh to 90 EUR/MWh, providing the lifetime of wind turbine increases from 20 years to 25 years and other significant cost reductions are achieved. Fig. 16 shows another example of U.S. estimated LCOE for several promising renewable energy technologies in 2018 [43]. It can be seen that the cost distribution of different technologies vary a lot, where the on shore wind power still shows cost advantages compared to the other renewable energy sources. It can be also expected that in the U.S. the capital cost may still be dominant for most of the renewable energy technologies for the next decade.

the voltage level increases. Also the weight of the whole converter system should be limited as much as possible because it has to be placed in the nacelle or tower of the turbines. The long cable connections between the nacelle and tower-base range from dozens to hundreds of meters and may require higher voltage level in the power conversion stage in order to reduce the cable losses and cable weight. Furthermore, remote locations of the wind turbines may increase the cost for installation and maintenance, which also demands high reliability, modularity and redundant design of the power electronics converter. B. Higher reliability The dramatic growth of total installation and also the individual capacity make the failures of wind turbines costly or even unacceptable from the point view of the TSO/grid owner. The failures of wind turbine system will not only cause stability problems of the power grid due to sudden absence of a large amount of power capacity, but also results in high cost for repairing and maintenance especially for those large and remote-located wind turbines, additionally it will cause loss of production of the owner leading to reduced total/annual energy production and thus increase the LCOE. As a result, the reliability performance is a critical design-consideration for the next generation wind power converters. When looking at the failure rates and down time distribution in individual wind turbine [45], it is found that the control and power electronic parts tend to have higher failure rate than the other subsystems with a factor of 2-4. That means the reliability improvements of power electronics will effectively reduce the cost of wind energy or LCOE. The reliability research in power electronics has been carried out for decades and now is moving from a solely statistical approach that has been proven to be unsatisfactory in the automotive industry, to a more physical-based approach which involves not only the statistics but also the investigation and modeling of the root cause behind the failures [46]-[48]. As shown in Fig. 17, a diagram of possible activities for reliability analysis and improvement is illustrated. In order to achieve more cost-efficient and reliable power electronic systems, multi-disciplinary approaches are necessary to be included which involve better mission profile mapping, stress analysis, strength modeling, statistical considerations, and also online monitoring/control of the converter for predictive maintenance. It is noted that a series of reliability design tools may be necessary in order to transfer, organize and combine the relevant approaches to more insight reliability performances. The stress analysis may focus on complete mission profile definition, converter design, stress estimation and measurement. This group of disciplines will target a better establishment of the converters loading profile, which can initiate the failure mechanisms of the critical components, like thermal cycling in power devices [49], voltage increase in the DC bus [50],vibration, humidity, and other stress factors [51], [52].

Fig. 16. Estimated Levelized Cost Of Energy for several renewable energy technologies in 2018. (Source: Energy Information Administration, Annual Energy Outlook 2013) [43].

As more power electronics are introduced in the energy system to improve the performances of power generation, the cost for the power electronics is becoming more important and it depends a lot on the specific wind turbine concepts. In the wind turbine system there are some special cost considerations, which impose challenges for the design and the selection of power electronics. For instance, the needs for higher power capacity and fullscale power conversion will increase the cost for power semiconductors, passive components and corresponding cooling management - thereby more efficient circuits and devices are required. Additionally, due to the limited space in the nacelle or tower, a higher power density will be emphasized for the power converters and this may lead to extra cost for insulation materials and also compact design as

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

C y c le stofa ilu r e

10

? Tj (K)

100

Fig. 17. Multi-disciplinary approaches for more reliable power electronics.

The strength modeling may involve the identification, modeling and accelerating test of failure mechanisms in the converter system, e.g. the bond wire lift-off and soldering crack inside the power devices [53]. The goal of this group of disciplines is to seek correlations between the established/measured stresses and the quantified fatigues/failures of the critical components. The monitoring and control approach may relate to the lifetime monitoring [54], [55], junction temperature control methods [56], [57] and intelligent maintenance using prognostic methods. This group of disciplines will target to monitor and control the converter lifetime during operation. For example, the Collector-Emitter voltage VCE of an IGBT, which is subject to an accelerated test, experiences a sudden increase just before the IGBT failure [54] and therefore can be used for predictive maintenance in the wind turbine system. The probability and statistics may add the statistical distribution and correlation to the acquired stress, strength and component configuration. This group of disciplines will target to enhance and validate the robustness of the designed converter and take into account the severe usage of wind turbine, six sigma strategies in the design and quality variations of components in order to give estimation about failure rate in the next 10 to 25 years of the product. With the better models and understanding of the reliability of power electronics, the LCOE of wind power may be significantly reduced because the design margins can be more accurately reserved, the loading of devices can be more intelligently modified, also the maintenance can be more wisely scheduled - achieving reduced cost for design and maintenance and extended lifetime for the wind turbines. C. More advanced grid integration features The grid codes nowadays reflect the trends of more penetration of wind energy into the power grid, and they just impose some basic requirements for the wind turbines to act more like the conventional power plants. It can be expected that in order to enable even larger-scale utilization of wind power in the future, more advanced grid interconnection requirements/features may be necessary not only for the wind turbine units but also for the wind farms or even the whole distribution networks including the wind turbines.

1) Protection and islanding operation

In order to reduce the impact of the wind gust and ensure the security and stability of grid operation, the wind turbines or wind farms are preferred to be configured as distributed generation networks (with many small-size power stations connected to the medium voltage distribution grid), whose power flows and electrical behaviors are different compared to the traditional centralized generation networks (with only a few large-sized power stations connected to the high voltage transmission grid). Therefore the protection schemes of the future grid utilities with more wind power penetration should be also changed resulting in a more distributed protection structure and may allow the islanding operation of some wind turbine units and local loads which are composed as microgrids. Such features demand more control efforts for the future wind turbine system in respect to the grid fault-identification, active and reactive power coordination, energy storage, signal communication, voltage stabilization. Moreover as the growing proportion of wind power in the power grid, more advanced grid requirements which usually are targeted to large power plants can also be addressed to the wind turbine itself. E.g. in the case of the faults and shutting down of transmission networks, the wind turbine systems may need the abilities to do black start if the faults are cleared (i.e. restart the power generation without any power supply from the grid). In order to achieve these more advanced features of grid interconnection, some energy storage systems may be needed for the future wind turbines and the wind farms. The storage system can be configured locally for each wind turbine unit, as shown in Fig. 18 (a), or be configured centrally for several wind turbines/wind farms, as shown in Fig. 18 (b). It is noted that the energy storage could be in the form of battery, supercapacitor, flywheel, hydro power station, or even combined energy storage systems, depending on the cost as well as the needed energy amount and control dynamics. Such wind turbine system with energy storage will also be ready to operate as primary controller in the case of enough energy is pre-stored and the wind power plant is approved to support this power system control feature.

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

2) Configurations of wind power plants

As the growing of wind power capacity, large wind farms which consist of many wind turbines are being developed. These wind farms may have significant impacts to the grids, and therefore they will play an important role on the power quality and the control of the power grid systems. The power electronic technology is again an important part for both the system configurations and the control of the wind farms in order to fulfil the growing grid demands. Some existing and potential configurations of the wind farms are shown in Fig. 19. A wind farm equipped with DFIG-based wind turbine systems is shown in Fig. 19 (a). Such a wind farm system is in operation in Denmark as a 160 MW off-shore wind power station. It is noted that due to the limitation of the reactive power capability, a centralized reactive power compensator like STATCOM may be needed in order to fully satisfy future grid requirements. Fig. 19 (b) shows another wind farm configuration equipped with a wind turbine system based on full-scale power converter. Because the reactive power controllability is significantly extended, the grid side converter in each of the generation unit can be used to provide the required reactive power individually, leading to reactive power compensatorless solutions. For long distance power transmission from an offshore wind farm, HVDC may be an interesting option because the efficiency is improved and no voltage compensators are needed [58], [59]. A typical HVDC transmission solution for wind power is shown in Fig. 19 (c), in which the medium ac

(a) Distributed energy storage.

(b) Centralized energy storage. Fig. 18. Potential energy storage configurations for wind power plants to enable virtual power plant operation.

(a)

(b)

(c)

(d)

Fig. 19. Potential wind farm configurations with AC and DC power delivery. (a) Doubly-fed induction generator system with ac-grid. (b) Full-scale converter system with ac-grid. (c) Full-scale converter system with transmission dc-grid. (d) Full-scale converter system with both distribution and transmission dc grid.

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

voltage of the wind farm output is converted into a high dc voltage by a boost transformer and high voltage rectifier. Another possible wind farm configuration with HVDC transmission is shown in Fig. 19 (d), where a solid state transformer (or DC/DC transformer) [60] is used to convert the low/medium DC voltage from each wind turbine output, to the medium/high DC voltage for transmission, thus a full DC power delivery both in the distribution and transmission grid can be realized. It is claimed in [61] that the overall efficiency of the power delivery can be significantly improved compared to the configuration shown in Fig. 19 (c) mainly because of less converters and transformers in this DC transmission system. It can be a future solution for large wind farms to increase the overall efficiency of power delivery. D. Future technologies for wind power conversion Nowadays most of the erected wind turbines are based on the power conversion at low voltage level (below 690 V) and using induction, permanent magnet, or wound rotor synchronous generators as discussed before. In order to satisfy the fast growing capacity and the tough challenges for the wind power system, advanced technologies which can realize more efficient and reliable wind power conversion are expected to appear in the near future, where the major changes could happen in the drive trains, generators, power converters and semiconductors devices, by achieving power conversion at higher voltage level (1 kV 10 kV).

(a)

(b)

1) Generator and drive train

Because of the experience for high robustness and relative cheap price in the motor drive applications, the high-speed induction/asynchronous generators are widely adopted in the wind turbine system in the past and even nowadays. However, in order to adapt the different rotational speeds between the wind turbine and generator, a multi-stage gear box which is normally heavy and bulky has to be used, as shown in Fig. 20 (a). The multi-stage gear box has been a trouble maker in the wind power industry due to the unsatisfactory down time as well as the high cost for maintenance [45]. Thereby, multipole low-speed synchronous generators were introduced in the wind power application because the rotational speed can be designed to adapt to the wind turbine speed even with no gear/direct-driven solution, as shown in Fig. 20 (c). However as the power level of the individual wind turbines keeps growing up even to 10 MW, it is becoming more and more difficult to get acceptable size and weight of the multi-pole generator at low rotational speed as the diameter is becoming huge. Actually in some of the newly presented wind turbines which range from 5 to 8 MW, a trade-off solution with 1-stage gear driven is seen to be applied, as shown in Fig. 20 (b), where the drive-train is simpler and smaller than the solution in Fig. 20 (a), meanwhile a medium speed synchronous generator can achieve smaller size than the generator solution in Fig. 20 (c). Nevertheless, the uncertain price for permanent magnet materials, which are popularly used in the existing multi-pole synchronous generators, may limit the concepts in Fig. 20 (b) and Fig. 20 (c) in order to be further commercialized calling for synchronous generator solutions with less/no permanent
(c) Fig. 20. Possible drivetrain and generation solutions for wind power system. (a) multi-stage gear drive with high speed generator (b) one-stage gear drive medium speed generator (c) direct drive with low speed generator.

magnets in the future. That trend also gives possibilities for new PM-less generator technologies such as high temperature superconductor (HTS) generator, which compared to conventional generators is significantly smaller and lighter with larger air gap to eliminate issues of tolerance or deformation making it very suitable to be utilized in the wind power application if regardless cost [62]. In respect to some other PM-less technologies like the low speed DFIG and switched reluctance generator, they do not seem to be suitable for multi-MW wind power conversion due to the requirements for small air gap which makes the turbine installation difficult.

2) Power electronics and transformer

Due to the space limitation in the nacelle or in the tower of the wind turbines, a full-scale converter solution which avoids the bulky low-frequency transformer may become a promising solution for the future wind turbine system. As shown in Fig. 21, this configuration shares the similar idea with the next generation traction converters [63], [64], and it is also proposed in the European UNIFLEX-PM Project [65]. It is based on a structure of back-to-back Cascaded H-bridge converter, with galvanic insulated DC/DC converters as

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

interface. The DC/DC converters have Medium Frequency Transformer (MFT) operating at several kHz to dozens of kHz, thereby the transformer size can be significantly reduced. Moreover, because of the cascaded structure, it can be directly connected to the distribution power grid (10 kV-33 kV) with high output quality, filter-less design, and also redundant ability which are of high interest in wind power system. This solution would become more attractive if it can be placed in the nacelle of wind turbines because the heavy/bulk low frequency transformer can be replaced by the more compact and flexible-configured power semiconductor devices. Another future configuration for wind turbines which shares the similar idea with some of the emerging converters used for High Voltage Direct Current (HVDC) Transmission [66], [67], is shown in Fig. 22. It is also based on a back-toback structure with cascaded converter cells of 2L-VSC. One advantage of this configuration is the easily scalable voltage/power capability; therefore it can achieve very high power conversion at dozens of kV with good modularity and redundancy. The output filter can also be eliminated because of significantly increased voltage levels. However, the useable voltage rating in the wind turbine system may be greatly limited by the insulation materials of the generator. Moreover, the low fundamental frequency of the generator outputs (which is the normal case for MW synchronous generator) may introduce large DC voltage fluctuation in the converter cells of the generator side, and thereby results in bulky capacitors in the converter system, which may be unpreferable for the wind power application.

3) Power semiconductor device

The power semiconductors are the backbone in the wind power converter and will determine many critical performances of the wind turbine system like the cost, efficiency, reliability, modularity. The potential high power silicon based semiconductor technologies in the wind power application are among the module packaged Insulated Gate Bipolar Transistor (IGBT), press-pack packaged IGBT and the press-pack packaging Integrated Gate Commutated Thyristor (IGCT) [68]-[70]. Recently there is a booming development of Silicon Carbide (SiC) based devices which are majorly in the form of Metal Oxide Semiconductor Field Effect Transistor (MOSFET) as well as diodes, although mainly targeted for low voltage and low power converter applications, they could be also the potential power devices used in wind power system. The four types of power semiconductor devices have quite different characteristics and they are compared in Table II. The module packaging technology of IGBT has a longer track record of applications and fewer mounting regulations. However, because of the soldering and bond-wire connection of internal chips, module packaging devices may suffer from larger thermal resistance, lower power density and higher failure rates [71]. The main trends to improve the packaging technology of IGBT module are to introduce pressure contact to eliminate the base plate and thus base plate soldering, sinter technology to avoid the chip soldering, as well as replaced bond wire material to reduce the coefcient of thermal expansion all lead to increased lifetime of module packaging IGBTs as reported in [72]. The press-pack packaging technology improves the connection of chips by direct press-pack contacting, which leads to improved reliability (yet to be scientifically proved but known from industrial experience), higher power density (easier stacking for series connection) and better cooling capability. Press-pack IGCTs were first introduced into the medium-voltage motor drivers in the 1990s and has already become state-of-the-art technology in the applications of oil, gas, HVDC, power quality, etc. However IGCT have not yet been mass adopted in the wind turbine system. As the power capacity of wind turbines grows up even to 10 MW, it can be expected that the press pack packaging devices may become more promising for the future wind turbine system. Besides the silicon power devices, the SiC based device which is claimed to have better switching characteristics and lower power losses, are also promising in the future wind power systems. Although the existing power capacity of the SiC devices are still not enough for the wind power application, these new device technologies show great potential for some future wind converter structures which consist of paralleled/cascaded converter cells, in which the requirements for voltage and current ratings are much lower. However the main challenges for SiC based devices are the bonding technology, limitation of stray inductance either in the external circuit or inside the module, higher dv/dt stress, higher operational temperature, as well as thinner chips all of these have to be carefully taken into account in the packaging and design of SiC based converter system.

Fig. 21. Cascade H-bridge converter with Medium Frequency Transformer for wind turbine.

Fig. 22. Modular multilevel converters for wind turbine.

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

Table II. Silicon power semiconductor devices for wind power application.

IGBT module
Power Density Reliability Cost Failure mode Easy maintenance Insulation of heat sink Snubber requirement Thermal resistance Switching loss Conduction loss Gate driver Major manufacturers Medium voltage ratings Max. current ratings

IGBT Press-pack High High High Short circuit

IGCT Press-pack High High High Short circuit

SiC MOSFET module Low Unknown High Open circuit

Low Moderate Moderate Open circuit

+ + Large Low Moderate Moderate Infineon, Semikron, Mitsubishi, ABB, Fuji 3.3 kV / 4.5 kV /6.5 kV 1.5 kV / 1.2 kA / 750 A

Small Moderate Moderate Moderate Westcode, ABB 2.5 kV / 4.5 kV 2.3 kA / 2.4 kA

+
Small Moderate Moderate Large ABB 4.5 kV /6.5 kV/10 kV 3.6 kA / 3.8 kA/ 2 kA

+ + Moderate Low Large Small Cree, Rohm, Mitsubishi 1.2 kV / 10 kV 180 A / 120 A

VII. CONCLUSION The total installed and individual capacity of wind turbines have both been steadily increasing in the last four decades mainly driven by the needs for more renewable energies and also constantly to lower the cost of energy. The wind power nowadays play much more important role in the energy supply system. The state-of-the-art configurations and roles of power electronics in the wind turbine system indicate that the behavior/performance of wind turbines can be significantly improved by introducing more advanced power electronic technologies. By proper controls and grid regulations, it is possible for the wind farms to act like conventional power plants and actively contribute to the frequency and voltage control in the grid system, thus the wind energy nowadays is more suitable to be integrated into the power grid. It is expected that even larger-scale integration of wind power will keep continuing in the near future. In the short term the main focus of the power electronics technology in wind power application will be on the grid integration and compatibility of the wind power in the power system. While in the long run, the challenges for lower energy cost, higher reliability, and better grid integration will push the technology developments in the wind turbine system and enable many possibilities in respect to the power delivery structure, drive train, generator, power electronics converter configurations and the semiconductor devices.
[1]

REFERENCES
M. Liserre, R. Cardenas, M. Molinas, J. Rodriguez, Overview of Multi-MW wind turbines and wind parks, IEEE Trans. on Industrial Electronics, Vol. 58, No. 4, pp. 1081-1095, April 2011. Z. Chen, J.M. Guerrero, F. Blaabjerg, "A Review of the State of the Art of Power Electronics for Wind Turbines," IEEE Trans. on Power Electronics, vol.24, No.8, pp.1859-1875, Aug 2009. F. Blaabjerg, Z. Chen, S.B. Kjaer, Power Electronics as Efficient Interface in Dispersed Power Generation Systems, IEEE Trans. on Power Electronics, Vol. 19, no. 4, pp. 11841194, 2004. A.D. Hansen, F. Iov, F. Blaabjerg, L.H. Hansen, Review of contemporary wind turbine concepts and their market penetration, Journal of Wind Engineering, Vol. 28, No. 3, pp. 247-263, 2004. M.P. Kazmierkowski, R. Krishnan, F. Blaabjerg, Control in Power Electronics-Selected problems, Academic Press, 2002. ISBN 0-12-402772-5. F. Blaabjerg, M. Liserre, K. Ma, Power Electronics Converters for Wind Turbine Systems, IEEE Trans. on Industry Application, vol. 48, no. 2, pp. 708-719, 2012. Website of Vestas Wind Power, Wind turbines overview, April 2011. (Available: http://www.vestas.com/). UpWind project, Design limits and solutions for very large wind turbines, March 2011, (Available: http://www.ewea.org/fileadmin/ewea_documents/documents/up wind/21895_UpWind_Report_low_web.pdf) REN21 - Renewables 2012 Global Status Report, June, 2012. (Available: http://www.ren21.net)

[2] [3]

[4]

[5] [6] [7] [8]

[9]

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

[10] Report of Danish Commission on Climate Change Policy, Green Energy - the road to a Danish energy system without fossil fuels, September 2010. (Available: http://www.klimakommissionen.dk/en-US/) [11] S. Muller, M. Deicke, R.W. De Doncker, "Doubly fed induction generator systems for wind turbines," IEEE Industry Applications Magazine, vol.8, no.3, pp.26-33, May/Jun 2002. [12] D. Xiang, Li Ran, P.J. Tavner, S. Yang, "Control of a doubly fed induction generator in a wind turbine during grid fault ridethrough," IEEE Trans. on Energy Conversion, Vol.21, no.3, pp. 652-662, Sept. 2006. [13] F.K.A Lima, A. Luna, P. Rodriguez, E. H. Watanabe, F. Blaabjerg, "Rotor Voltage Dynamics in the Doubly Fed Induction Generator During Grid Faults," IEEE Trans. on Power Electronics, vol.25, no.1, pp.118-130, Jan. 2010. [14] D. Santos-Martin, J.L. Rodriguez-Amenedo, S. Arnaltes, "Providing Ride-Through Capability to a Doubly Fed Induction Generator Under Unbalanced Voltage Dips," IEEE Trans. on Power Electronics, vol.24, no.7, pp.1747-1757, July 2009. [15] R. Pena, J.C. Clare, G.M. Asher, Doubly fed induction generator using back-to-back PWM converters and its application to variable speed wind-energy generation, Electric Power Application, Vol. 143, No. 3, 1996, pp. 231-241. [16] B. Andresen, J. Birk,A high power density converter system for the Gamesa G10x 4.5 MW Wind turbine, in Proc. of EPE2007, pp. 1-7, 2007. [17] R. Jones, P. Waite, Optimised power converter for multi-MW direct drive permanent magnet wind turbines, in Proc. of EPE2011, pp.1- 10, 2011. [18] J. Rodriguez, S. Bernet, Wu Bin, J. O. Pontt, S. Kouro, Multilevel Voltage-Source-Converter Topologies for Industrial Medium-Voltage Drives, IEEE Trans. on Industrial Electronics, vol. 54, no. 6, pp. 2930-2945, 2007. [19] S. Kouro, M. Malinowski, K. Gopakumar, J. Pou, L. G. Franquelo, B. Wu, J. Rodriguez, M. A. Perez, J. I. Leon, Recent Advances and Industrial Applications of Multilevel Converters, IEEE Trans.on Power Electronics, vol. 57, no. 8, pp. 2553 2580, 2010. [20] A. Faulstich, J.K. Stinke, F. Wittwer, Medium voltage converter for permanent magnet wind power generators up to 5 MW, in Proc. EPE 2005, pp.1- 9, 2005. [21] N. Celanovic, D. Boroyevich, A comprehensive study of neutral-point voltage balancing problem in three-level neutralpoint-clamped voltage source PWM inverters, IEEE Trans. on Power Electronics, 2000, vol. 15, no. 2, pp. 242-249. [22] S. Srikanthan, M.K. Mishra, "DC Capacitor Voltage Equalization in Neutral Clamped Inverters for DSTATCOM Application ," IEEE Trans. on Industrial Electronics, vol. 57, no. 8, pp. 2768-2775, August 2010. [23] J. Zaragoza, J. Pou, S. Ceballos, E. Robles, C. Jaen, M. Corbalan, "Voltage-Balance Compensator for a Carrier-Based Modulation in the Neutral-Point-Clamped Converter," IEEE Trans. on Industrial Electronics, vol. 56, no. 2, pp. 305-314, Feb 2009. [24] K. Ma, F. Blaabjerg, D. Xu, Power Devices Loading in Multilevel Converters for 10 MW Wind Turbines, in Proc. of ISIE 2011, pp. 340-346, June 2011. [25] K. Ma, F. Blaabjerg, Multilevel Converters for 10 MW Wind Turbines, in Proc. of EPE2011, pp. 1-10, 2011 Birmingham. [26] J. Rodriguez, S. Bernet, P. K. Steimer, I. E. Lizama, A Survey on Neutral-Point-Clamped Inverters, IEEE Trans. on Industrial Electronics, vol. 57, no. 7, pp. 2219-2230, 2010. [27] M. S. El-Moursi, B. Bak-Jensen, M.H. Abdel-Rahman, "Novel STATCOM Controller for Mitigating SSR and Damping Power System Oscillations in a Series Compensated Wind Park,"

[28]

[29]

[30]

[31]

[32]

[33]

[34] [35]

[36]

[37] [38] [39] [40] [41]

[42] [43]

[44] [45] [46]

IEEE Trans. on Power Electronics, vol.25, no.2, pp.429-441, Feb. 2010. J. Dai, D.D. Xu, B. Wu, "A Novel Control Scheme for CurrentSource-Converter-Based PMSG Wind Energy Conversion Systems," IEEE Trans. on Power Electronics, vol.24, no.4, pp.963-972, April 2009. X. Yuan, F. Wang, D. Boroyevich, Y. Li, R. Burgos, "DC-link Voltage Control of a Full Power Converter for Wind Generator Operating in Weak-Grid Systems," IEEE Trans. on Power Electronics, vol.24, no.9, pp.2178-2192, Sept. 2009. P. Rodriguez, A. Timbus, R. Teodorescu, M. Liserre, F. Blaabjerg, "Reactive Power Control for Improving Wind Turbine System Behavior Under Grid Faults," IEEE Trans. on Power Electronics, vol.24, no.7, pp.1798-1801, July 2009. A.Timbus, M. Liserre, R. Teodorescu, P. Rodriguez, F. Blaabjerg, "Evaluation of Current Controllers for Distributed Power Generation Systems," IEEE Trans. on Power Electronics, vol.24, no.3, pp.654-664, March 2009. M. Liserre, F. Blaabjerg, S. Hansen, "Design and control of an LCL-filter-based three-phase active rectifier," IEEE Trans. on Industry Applications, vol.41, no.5, pp. 1281- 1291, Sept.-Oct. 2005. P. Rodriguez, A.V. Timbus, R. Teodorescu, M. Liserre, F. Blaabjerg, "Flexible Active Power Control of Distributed Power Generation Systems During Grid Faults," IEEE Trans. on Industrial Electronics, vol.54, no.5, pp.2583-2592, Oct. 2007. R. Teodorescu, M. Liserre and P. Rodriguez, Grid Converters for Photovoltaic and Wind Power Systems. IEEE/Wiley, 2011. F. Blaabjerg, R. Teodorescu, M. Liserre, A.V. Timbus, "Overview of Control and Grid Synchronization for Distributed Power Generation Systems," IEEE Trans. on Industrial Electronics, vol.53, no.5, pp.1398-1409, Oct. 2006. M. Altin, O. Goksu, R. Teodorescu, P. Rodriguez, B. BakJensen, L. Helle, Overview of recent grid codes for wind power integration, in Proc. of OPTIM2010, pp.1152-1160, 2010. M. Tsili, A review of grid code technical requirements for wind farms, IET Journal of Renewable Power Generation, Vol.3, no.3, pp. 308-332, 2009. Energinet Wind turbines connected to grids with voltages below 100 kV, Jan. 2003. Energinet Technical regulation 3.2.5 for wind power plants with a power output greater than 11 kW, Sep. 2010. E.ON-Netz Grid Code. Requirements for offshore grid connections in the E.ON Netz network, April 2008. Wikipedia Cost of electricity by source, April 2013. (Avaiable: http://en.wikipedia.org/wiki/Cost_of_electricity_by_source). Denmark-Supplier of Competitive Offshore Wind Solutions. Megavind's Strategy for Offshore Wind Research, Development and Demonstration, 2010. Report of the US Energy Information Administration (EIA) of the U.S. Department of Energy (DOE). Levelized Cost of New Generation Resources in the Annual Energy Outlook 2013, Released in spring, 2013. S. Faulstich, P. Lyding, B. Hahn, P. Tavner Reliability of offshore turbinesidentifying the risk by onshore experience, in Proc. of European Offshore Wind, Stockholm, 2009. B. Hahn, M. Durstewitz, K. Rohrig Reliability of wind turbines Experience of 15 years with 1500 WTs, Wind Energy, Spinger, Berlin, 2007. E. Wolfgang, L. Amigues, N. Seliger and G. Lugert, Buildingin Reliability into Power Electronics Systems. The World of Electronic Packaging and System Integration, pp. 246-252, 2005.

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

[47] D. Hirschmann, D. Tissen, S. Schroder, R.W. De Doncker, "Inverter design for hybrid electrical vehicles considering mission profiles," IEEE Conference on Vehicle Power and Propulsion, Sept. 2005. [48] H. Wang, M. Liserre, F. Blaabjerg, "Toward reliable power electronics - challenges, design tools and opportunities," IEEE Industrial Electronics Magazine, vol. 7, no. 2, pp. 17-26, Jun. 2013. [49] C. Busca, R. Teodorescu, F. Blaabjerg, S. Munk-Nielsen, L. Helle, T. Abeyasekera, P. Rodriguez, An overview of the reliability prediction related aspects of high power IGBTs in wind power applications, Microelectronics Reliability, Vol. 51, no. 9-11, pp. 1903-1907, 2011. [50] N. Kaminski, A. Kopta, Failure rates of HiPak Modules due to cosmic rays, ABB application note 5SYA 2042-04, Mar 2011. [51] E. Wolfgang, Examples for failures in power electronics systems, presented at ECPE Tutorial on Reliability of Power Electronic Systems, Nuremberg, Germany, April 2007. [52] S. Yang, A. T. Bryant, P. A. Mawby, D. Xiang, L. Ran, and P. Tavner, An industry-based survey of reliability in power electronic converters, IEEE Trans. on Ind. Appl., vol. 47, no. 3, pp. 1441- 1451, May/Jun. 2011. [53] ABB Application Note, Load-cycling capability of HiPak IGBT modules, 2012. [54] S. Yang, D. Xiang, A. Bryant, P. Mawby, L. Ran, P. Tavner, "Condition Monitoring for Device Reliability in Power Electronic Converters: A Review," IEEE Trans. on Power Electronics, vol.25, no.11, pp.2734-2752, Nov. 2010. [55] J. Due, S. Munk-Nielsen, R. Nielsen, Lifetime investigation of high power IGBT modules, in Proc. of EPE2011 Birmingham, 2011. [56] K. Ma, F. Blaabjerg, Thermal Optimized Modulation Method of Three-level NPC Inverter for 10 MW Wind Turbines under Low Voltage Ride Through, IET Journal on Power Electronics, Vol. 5, No. 6, pp. 920-927, 2012. [57] K. Ma, M. Liserre, F. Blaabjerg, Reactive Power Influence on the Thermal Cycling of Multi-MW Wind Power Inverter, IEEE Trans. on Industry Applications, vol. 49, no. 2, pp. 922930, 2013. [58] A. Prasai, Y. Jung-Sik, D. Divan, A. Bendre, Seung-Ki Sul, "A New Architecture for Offshore Wind Farms," IEEE Trans. on Power Electronics, vol.23, no.3, pp.1198-1204, May 2008. [59] F. Iov, P. Soerensen, A. Hansen, F. Blaabjerg, Modelling, Analysis and Control of DC-connected Wind Farms to Grid, International Review of Electrical Engineering, Praise Worthy Prize, February 2006, pp.10, ISSN 1827-6600. [60] S. Engel, N. Soltau, H. Stagge, R. W. De Doncker Dynamic and Balanced Control of Three-Phase High-Power Dual-Active Bridge DC-DC Converters in DC-Grid Applications, IEEE Trans. on Power Electronics, vol. 28, no. 4, pp. 1880-1889, 2012. [61] C. Meyer, M. Hoing, A. Peterson, and R. De Doncker, Control and design of dc grids for offshore wind farms, IEEE Trans. Ind. Appl., vol. 43, no. 6, pp. 14751482, Nov./Dec. 2007. [62] G. Snitchler, B. Gamble, C. King, P. Winn, 10 MW Class Superconductor Wind Turbine Generators, IEEE Trans. on Applied Superconductivity, vol. 21, no. 3, pp. 1089-1092, 2011. [63] B. Engel, M. Victor, G. Bachmann, and A. Falk, 15 kV/16.7 Hz energy supply system with medium frequency transformer and 6.5 kV IGBTs in resonant operation, in Proc. of EPE2003, pp. 1-10, Sep. 2003. [64] S. Inoue, H. Akagi, A Bidirectional Isolated DCDC Converter as a Core Circuit of the Next-Generation MediumVoltage Power Conversion System, IEEE Trans. on Power Electronics, vol.22, no. 2, pp. 535-542, 2007.

[65] F. Iov, F. Blaabjerg, J. Clare, O. Wheeler, A.Rufer, A. Hyde, UNIFLEX-PM-A Key-Enabling Technology for Future European Electricity Networks, EPE Journal, vol. 19, no. 4, pp.6-16, 2009. [66] M. Davies, M. Dommaschk, J. Dorn, J.Lang, D. Retzmann, D. Soerangr, HVDC PLUS Basics and Principles of Operation, Siemens Technical articles, 2008. [67] A. Lesnicar and R. Marquardt, An innovative modular multilevel converter topology suitable for a wide power range, in Proc. of IEEE Bologna PowerTech Conf., pp. 16, 2003. [68] K. Ma, F. Blaabjerg, The Impact of Power Switching Devices on the Thermal Performance of a 10 MW Wind Power NPC Converter, Energies 5, no. 7: 2559-2577. [69] R. Jakob, C. Keller, B. Gollentz, 3-Level high power converter with press pack IGBT, in Proc. of EPE 2007, pp. 2-5, Sept. 2007. [70] R. Alvarez, F. Filsecker, S. Bernet, Comparison of press-pack IGBT at hard switching and clamp operation for medium voltage converters, in Proc. of EPE2011, pp. 1-10, 2011. [71] U. Scheuermann, Reliability challenges of automotive power electronics, Microelectronics Reliability, vol. 49, no. 9-11, pp. 1319-1325, 2009. [72] U. Scheuermann, Ralf Schmidt, A New Lifetime Model for Advanced Power Modules with Sintered Chips and Optimized Al Wire Bonds, Proc. of PCIM 2013, pp. 810-813, 2013.

Frede Blaabjerg (S86-M88SM97-F03) He was employed at ABB-Scandia, Randers, from 1987-1988. During 1988-1992 he was PhD. student at Aalborg University, Denmark, became Assistant Professor in 1992, Associate Professor in 1996 and full professor in power electronics and drives in 1998. He has been parttime research leader at Research Center Risoe in wind turbines. In 2006-2010 he was dean of the faculty of Engineering, Science and Medicine and became visiting professor at Zhejiang University, China in 2009. His research areas are in power electronics and its applications like in wind turbines, PV systems and adjustable speed drives. He has been Editor in Chief of the IEEE Transactions on Power Electronics 20062012. He was Distinguished lecturer for the IEEE Power Electronics Society 2005-2007 and for IEEE Industry Applications Society from 2010-2011. He has been Chairman of of EPE2007 and PEDG2012 both held in Aalborg. He received the 1995 Angelos Award for his contribution in modulation technique and the Annual Teacher prize at Aalborg University. In 1998 he received the Outstanding Young Power Electronics Engineer Award from the IEEE Power Electronics Society. He has received thirteen IEEE Prize paper awards and another prize paper award at PELINCEC Poland 2005. He received the IEEE PELS Distinguished Service Award in 2009 and the EPE-PEMC 2010 Council award. Finally he has received a number of major research awards in Denmark.

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.

This article has been accepted for publication in a future issue of this journal, but has not been fully edited. Content may change prior to final publication.

Ke Ma (S09-M11) He received the B.Sc. and M.Sc. degrees in electrical engineering from the Zhejiang University, Hangzhou, China in 2007 and 2010 respectively. He is currently working toward the Ph.D. degree in the Department of Energy Technology, Aalborg University, Aalborg, Denmark. His research interests are in the topologies, control, modulation and reliability in the wind power generation system. He has received the outstanding presentation award at APEC 2013, the IEEE Industry Applications Society third prize paper award in 2012, and a prize paper award at ISIE 2011.

Copyright (c) 2013 IEEE. Personal use is permitted. For any other purposes, permission must be obtained from the IEEE by emailing pubs-permissions@ieee.org.